Title :
Feature selection methods of the combined feature vector for classifying diffuse lung opacities in thin section computed tomography
Author :
Mitani, Yoshihiro ; Fujita, Yusuke ; Matsunaga, Naofumi ; Hamamoto, Yoshihiko
Author_Institution :
Yamaguchi Junior Coll., Japan
Abstract :
In designing a computer-aided diagnosis (CAD) system for classifying diffuse lung opacities in thin-section computed tomography (HRCT) images, the computational cost is an important factor. In this paper, the feature selection methods for reducing the dimensionality of the combined feature vector are investigated.
Keywords :
computer aided analysis; computerised tomography; feature extraction; image reconstruction; lung; medical image processing; CAD system; Fisher criterion; Mahalanobis distance; combined feature vector; computational cost; computer-aided diagnosis system; diffuse lung opacities classification; feature selection methods; sequential forward floating selection method; sequential forward selection; thin section computed tomography; Computational efficiency; Computed tomography; Computer aided diagnosis; Coronary arteriosclerosis; Design automation; Diseases; Educational institutions; Gabor filters; Histograms; Lungs;
Conference_Titel :
Biomedical Engineering, 2003. IEEE EMBS Asian-Pacific Conference on
Print_ISBN :
0-7803-7943-8
DOI :
10.1109/APBME.2003.1302657
